RTTV-TCP: Adaptive congestion control algorithm based on RTT variations for mmWave networks

• Main Authors: Alramli, Omar Imhemed, Mohd Hanapi, Zurina, Othman, Mohamed, Ahmad, Idawaty, Samian, Normalia
• Format: Article
• Language: English
• Published: Elsevier 2024
• Online Access: http://psasir.upm.edu.my/id/eprint/113677/1/113677.pdf; http://psasir.upm.edu.my/id/eprint/113677/; https://www.sciencedirect.com/science/article/pii/S1570870524002221

Internet applications such as video gaming virtual/ augmented reality necessitate efficient fifth-generation (5G) millimeter-wave (mmWave) cellular networks. The Transmission Control Protocol (TCP) is an essential protocol for network connectivity. However, TCP faces challenges in efficiently utilizing the available bandwidth of 5G mmWave cellular networks while maintaining low latency, mainly due to constraints like Non-Line of Sight (NLoS) conditions. This paper introduces Round-Trip-Time Variations-TCP (RTTV-TCP), enhancing TCP performance in 5G mmWave cellular networks. Simulation scenarios for a 5G mmWave cellular network have been conducted to evaluate RTTV-TCP's performance, comparing it to legacy TCP variants such as NewReno, HighSpeed, CUBIC, Bottleneck Bandwidth and Round-trip propagation time (BBR), FB-TCP (Fuzzy Based-TCP). The results demonstrate that RTTV-TCP achieves higher average throughput than these TCP variants while maintaining the same level of delay in 5G mmWave cellular networks. RTTV-TCP outperforms NewReno and CUBIC by a very significant margin, demonstrating a 208% improvement compared to HighSpeed and a 6% increase compared to BBR protocol in the worst Packet Error Rate (PER) scenario and when the buffer size matches the Bandwidth Delay Product (BDP). © 2024 Elsevier B.V.